Please help put the below paragraphs into more concise wording, thank you very much## Introduction

Voter opinions on politics, party and party leader are essential for predicting the development of party. Through analyzing voter opinions, parties can find out the real situation and forecast the results of the next round of general election. Knowing the real needs and opinions of the people, it will help them gain more supporters by satisfying the people's needs and wants. In the end, this all comes down to the result of winning more votes and winning the election.

The data used in the project will be the 2019 Online Canadian Election Survey data. Based on this data, we will be providing insights to the Liberal Party about voter opinions specifically on immigrants, education, and employment to make data-based recommendations.

## Research Question 1

For the first research question, we would like to study that among all the Canadian citizens who are older than 18 and rank the Liberal Party as their first choice, is there an association between the rating of Liberal Party and the rating of Immigrants? We would further want to explore this question in depth by exploring the strength of the association, and the direction of the association.

Motivation: Liberal Party is known to be very supportive of the intake of immigrants. However, it is essential to know whether voters have a positive or negative rating about immigrants, which can directly affect their decisions on whether they would vote for a party that is a huge supporter of admitting immigrants into the country.

Data Summary: We used the 2019 election survey data. For this research question, we first filtered the data by all participants in the survey that are above 18 years old and those who are Canadian Citizens. Then we filtered out all the NA values to make sure we would not get NA results when performing the linear regression.

## Statistical Method

We used simple linear regression to investigate whether there is association between the rating of Liberal Party and immigrants. The variables used for this method are groups_theorem_2 and party_rating_23.

1. First, we calculated the correlation value to have a basic idea of the strength and direction of the association.

2. Divided the sample into training and testing data and fitted the linear regression line based on the training data.

3. We stated the hypothesis. For this research question, the null hypothesis would be that there is no linear association between the rating of Liberal Party and the rating of immigrants. The alternative hypothesis would be that there exists a linear association between the rating of Liberal Party and the rating of immigrants.

4. Produced a table of summary to know the intercept, slope, and p-value based on the fitted regression line.

5. Lastly, we draw the conclusion of the research question, where we compare the p-value with the predetermined level to conclude whether there is a linear association between the rating of Liberal Party and the rating of Immigrants.

##Results

From the result from our model, we can see a p-value of 0, which is less than the pre-determined confidence interval 0.05, therefore, we can conclude that we have strong evidence against the null hypothesis which states that there is no association between the rating of Liberal Party and the rating of immigrants. The equation of the fitted linear regression line is $$text{Rating of Immigrants}= 21.7242 + text{Rating of Liberal Party}*0.4557$$ This result depicts that we have a positive slope for our association between Liberal Party and immigrants, which means that it is expected that the higher rating we have for immigrants, the higher the rating for Liberal Party. That is, there should be an 0.3144 increase in the rating of Immigrants for every one rating point higher in Liberal Party.

## Limitations

In this research, we were only limited to the data of voter opinions in 2019, which is not enough to investigate and predict the overall voters' opinions. Many factors can between 2019 and the next round of election. For example, this year, we have the Covid-19, which can become a severe changing factor for the next round of election. Since people would have experienced hardships such as difficulty of finding a job or losing a job, attending schools online, and facing financial hardships, voters' opinions on the election will very likely to change after this series of event. Another limitation in the project is in research question #3, where we created the new variable "employment3" by grouping the voters by employed and unemployed. However, there were a few arbitrary selections such as "Students" and "Disabled" that we cannot determine whether they are employed or unemployed. Thus, it is possible that the results of the recommendations could be influenced by these limitations.

## Conclusion

Based on all the researches we did above, we can first conclude that there is a positive association between the rating of Liberal Party and rating of immigrants, which means that people who vote for Liberal Party are more likely to accept immigrants, thus, those that support the intake of immigrants can possibly be the target people to promote towards for the next election.

Furthermore, from the second research question, many more people wanted to see more spending in education. Thus, it can suggest that for the next election, Liberal Party can possibly consider more on the funding and improvement of education systems in Canada to win the hearts of those who want to see more spending in education, which is eventually the majority of voters.

Lastly, from the last research question, we can conclude that the proportion of employment between Liberal Party, and its main competitor, Conservative Party, are equal. Thus, it can possibly suggest that employment may not be one of the potential influencers. Moreover, since the proportion of employment rate is relatively low (~0.6 for both parties), Liberal Party can possibly consider bringing more policies for better employment benefits that can appeal to the voters.

pur-new-sol

Related Questions